Predicting depression from neuroelectric data

Abstract

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for causing a stimulus presentation system to present first content to a patient. Obtaining, from a brainwave sensor, electroencephalography (EEG) signals of the patient while the first content is being presented to the patient. Identifying, from within the EEG signals of the patient, first brainwave signals associated with a first brain system of the patient, the first brainwave signals representing a response by the patient to the first content. Determining, based on providing the first brainwave signals as input features to a machine learning model, a likelihood that the patient will experience a type of depression within a period of time. Providing, for display on a user computing device, data indicating the likelihood that the patient will experience the type of depression within the period of time.

1 . A depression prediction system, comprising:
 one or more processors;   one or more tangible, non-transitory media operably connectable to the one or more processors and storing instructions that, when executed, cause the one or more processors to perform operations comprising:
 causing a stimulus presentation system to present first content to a patient, the first content including interactive content configured to test the patient's responses to receiving rewards and taking risks; 
 obtaining, from a brainwave sensor, electroencephalography (EEG) signals of the patient while the first content is being presented to the patient; 
 identifying, from within the EEG signals of the patient, first brainwave signals associated with a dopaminergic brain system of the patient, the first brainwave signals representing a response by the patient to the first content; 
 causing the stimulus presentation system to present second content to the patient, the second content being different from the first content, the second content including a sequence of images representing positive, neutral, and negative emotional stimuli; 
 obtaining EEG signals of the patient while the second content is being presented to the patient; and 
 identifying, from within the EEG signals of the patient, second brainwave signals associated with the amygdala and greater emotion processing system of the patient, the second brainwave signals representing a response by the patient to the second content, 
 determining, based on providing the first brainwave signals and the second brainwave signals as input features to a machine learning model, a likelihood that the patient will experience a type of depression within a period of time; and 
 providing, for display on a user computing device, data indicating the likelihood that the patient will experience the type of depression within the period of time. 
   
     
     
         2 . The system of  claim 1 , wherein determining the likelihood that the patient will experience the type of depression within the period of time comprises determining a severity of the type of depression. 
     
     
         3 . The system of  claim 1 , wherein the machine learning model is a convolutional neural network. 
     
     
         4 . The system of  claim 1 , wherein the machine learning model is a supervised machine learning model configured to be adaptive to actual patient diagnoses of depression. 
     
     
         5 . The system of  claim 1 , wherein the machine learning model is trained based on more than one hundred data sets of clinical test data. 
     
     
         6 . A computer-implemented depression prediction method executed by one or more processors and comprising:
 causing, by the one or more processors, a stimulus presentation system to present first content to a patient;   obtaining, by the one or more processors and from a brainwave sensor, electroencephalography (EEG) signals of the patient while the first content is being presented to the patient;   identifying, by the one or more processors and from within the EEG signals of the patient, first brainwave signals associated with a first brain system of the patient, the first brainwave signals representing a response by the patient to the first content;   determining, based on providing the first brainwave signals as input features to a machine learning model, a likelihood that the patient will experience a type of depression within a period of time; and   providing, for display on a user computing device, data indicating the likelihood that the patient will experience the type of depression within the period of time.   
     
     
         7 . The method of  claim 6 , wherein the first content is selected to trigger a response by a particular brain system of the patient. 
     
     
         8 . The method of  claim 6 , wherein the first brain system is a dopaminergic system or amygdala emotional system. 
     
     
         9 . The method of  claim 6 , further comprising:
 causing the stimulus presentation system to present second content to the patient, the second content being different from the first content;   obtaining EEG signals of the patient while the second content is being presented to the patient; and   identifying, from within the EEG signals of the patient, second brainwave signals associated with a second brain system of the patient, the second brainwave signals representing a response by the patient to the second content,   wherein determining the likelihood that the patient will experience the type of depression within the period of time comprises determining, based on providing the first brainwave signals and the second brainwave signals as input features to the machine learning model, the likelihood that the patient will experience the type of depression within the period of time.   
     
     
         10 . The method of  claim 9 , wherein the first brain system is a reward system and the second brain system is an emotion system. 
     
     
         11 . The method of  claim 9 , further comprising:
 obtaining EEG signals of the patient while no content is presented to the patient; and   identifying, from within the EEG signals of the patient, third brainwave signals associated with a resting state of the patient,   wherein determining the likelihood that the patient will experience the type of depression within the period of time comprises determining the likelihood that the patient will experience the type of depression within the period of time based on the first brainwave signals, the second brainwave signals, and the third brainwave signals.   
     
     
         12 . The method of  claim 6 , wherein determining the likelihood that the patient will experience the type of depression within the period of time comprises determining a severity of the type of depression. 
     
     
         13 . The method of  claim 6 , wherein the machine learning model is a convolutional neural network. 
     
     
         14 . The method of  claim 6 , wherein the machine learning model is a supervised machine learning model configured to be adaptive to actual patient diagnoses of depression. 
     
     
         15 . The method of  claim 6 , wherein the machine learning model is trained based on more than one hundred data sets of clinical test data. 
     
     
         16 . The method of  claim 6 , wherein the type of depression includes major depressive disorder or post-partum depression. 
     
     
         17 . The method of  claim 6 , wherein the first content includes interactive content configured to test the patient's responses to receiving rewards and taking risks. 
     
     
         18 . The method of  claim 6 , wherein the first content includes a sequence of images representing positive, neutral, and negative emotional stimuli. 
     
     
         19 . A non-transitory computer readable storage medium storing instructions that, when executed by at least one processor, cause the at least one processor to perform operations comprising:
 causing a stimulus presentation system to present first content to a patient;   obtaining, by the one or more processors and from a brainwave sensor, electroencephalography (EEG) signals of the patient while the first content is being presented to the patient;   identifying, from within the EEG signals of the patient, first brainwave signals associated with a first brain system of the patient, the first brainwave signals representing a response by the patient to the first content;   determining, based on providing the first brainwave signals as input features to a machine learning model, a likelihood that the patient will experience a type of depression within a period of time; and   providing, for display on a user computing device, data indicating the likelihood that the patient will experience the type of depression within the period of time.   
     
     
         20 . The medium of  claim 19 , wherein determining the likelihood that the patient will experience the type of depression within the period of time comprises determining a severity of the type of depression.